Changes in daily substance use among people experiencing homelessness and mental illness: 24-month outcomes following randomization to Housing First or usual care

Aims Housing First (HF) is an established intervention for people experiencing homelessness and mental illness. We compared daily substance use (DSU) between HF and treatment as usual (TAU). Design Two concurrent randomized controlled trials with 24‐month follow‐up. Setting Market rental apartments with support provided by Assertive Community Treatment (ACT) or Intensive Case Management (ICM); a single building with on‐site supports (CONG); TAU in Vancouver, Canada. Participants Inclusion criteria were current homelessness and mental illness. Participants were assessed as having either ‘high needs’ (HN; n = 297) or ‘moderate needs’ (MN; n = 200). MN participants were randomized to ICM (n = 100) or MN‐TAU (n = 100). HN participants were randomized to ACT (n = 90), CONG (n = 107) or HN‐TAU (n = 100). Interventions and comparators All HF interventions included independent housing with support services, with an emphasis on promoting client choice and harm reduction in relation to substance use. TAU included existing services and support available to homeless adults with mental illness. Measurements DSU over 24 and 12 months was derived from the Maudsley Addiction Profile. Also measured were demographics, homelessness history, psychiatric diagnoses, symptom severity, comorbid illnesses and duration of stable housing. Findings Compared with HN‐TAU, neither CONG [adjusted odds (AOR) ratio = 0.73, 95% confidence interval (CI) = 0.39–1.37] nor ACT (AOR = 1.22, 95% CI = 0.61–2.45) differed on DSU at 24 months, and MN‐TAU did not differ from ICM (AOR = 0.78, 95% CI = 0.37–1.63). There were no differences at 12 months, when analyses were restricted to participants who indicated substance use at baseline, or when considering the duration of stable housing. Conclusions Housing First, an intervention to support recovery for homeless people who have co‐occurring mental illness and substance use disorders, did not reduce daily substance use compared with treatment as usual after 12 or 24 months.